銀行業會計穩健性衡量變數之研究

Abstract

這篇論文主要在研究銀行業三個期初會計穩健性衡量變數對於盈餘反應好壞消息的影響，三個衡量變數分別為:(1)不良資產覆蓋率:備抵壞帳對不良資產之比率，以ALL/NPA表示，(2)備抵壞帳對淨壞帳沖銷數之比率，以ALL/NLWO表示及(3)市場價值對帳面價值之比率，以MTB表示。 實證結果發現，期初變數ALL/NPA及ALL/NLWO，於當期盈餘對好消息之反應呈正相關，而對壞消息之反應呈負相關，該結果支持「會計資訊受到穩健性的影響且盈餘操縱的影響並不顯著」情境。而期初變數MTB則於當期盈餘對好消息及壞消息之反應皆呈現負相關，該結果支持「會計資訊受到盈餘操縱的影響並不顯著」情境。因此推論1987-2004年美國金融控股公司平均而言會計資訊受到穩健性的影響且盈餘操縱的影響並不顯著。 但亦發現相較於2001-2004期間，在1987-1990及1991-2000這二個期間，會計資訊有受到較明顯的盈餘操縱影響；此外，大規模公司及高資本適足率公司其盈餘操縱的行為亦較小規模及低資本適足率公司顯著。

This paper is to investigate the impact of three conservatism measures in the banking industry on the association between earnings and stock returns in good and bad news. Three conservatism measures being tested are (1) the NPA coverage ratio: the ratio of allowance for loan losses to total non-performing assets (ALL/NPA), (2) the ratio of allowance for loan and lease losses to net loan write-off (ALL/NLWO), and (3) the market-to-book ratio (MTB). We find that conditional measures, beginning-of-period ALL/NPA and ALL/NLWO are positively related to earning timeliness to good news but have negative association with earning timeliness to bad news consistent with the scenario of “conservatism with no/non-intensive manipulation” on average. Also, the negative association between (1) unconditional measures, beginning-of-period MTB and (2) timeliness and asymmetry of recognition is verified in our sample of bank holding companies. Furthermore, we find that on average manipulation behavior is consistently more prevalent in bank holding companies over period 1987-1990 and 1991-2000 as comparison to period 2001-2004. Besides, big-scale firms and high capital ratio firms, as comparison to small-scale firms and low capital ratio firms, tend to engage in heavier manipulation over particular poor (1987-1990) and boom (1991-2000) time.